We frequently see this alert on our cluster
Metrics Collector ProcessTCP OK - 0.001s response on port 6188
CRITICALMetrics Collector ProcessConnection failed: timed out to 0.0.0.0:6188
Metrics collecotr goign down and coming back very fast. this is generating lot of alerts. is there a way to fix this?
@ARUN change the value of timeline.metrics.service.webapp.address from 0.0.0.0:6188 to <Ambari_Metrics_Collector_Server>:6188 and also check for the memory settings of Ambari Metrics.